Lakera·3 hours agoYou shall not pass: the spells behind GandalfThis article was originally posted on our company website. Lakera’s developer platform enables ML teams to ship fail-safe computer vision models. Spoiler alert: the article below contains some of the actual passwords! If you want to give Gandalf a try first, go here. Gandalf has captivated the world and has…Ai Safety10 min readAi Safety10 min read
Cory Doctorow·4 days agoMember-onlyAyyyyyy EyeeeeeThe lie that raced around the world before the truth got its boots on. — Tomorrow (June 5) at 7:15PM, I’m in London at the British Library with my novel Red Team Blues, hosted by Baroness Martha Lane Fox. On Tuesday (June 6), I’m on a Rightscon panel about interoperability. On Wednesday (June 7), I’m keynoting the Re:publica conference in Berlin. On Thursday (June 8)…Ai Safety10 min readAi Safety10 min read
TechGig·5 hours agoMarc Andreessen dispels AI myths: It’s not there to kill youDescribing AI as the application of mathematics and software code to teach computers how to understand and generate knowledge similarly to humans, Andreessen highlights that AI’s training revolves around human language and identifying patterns within that data. Venture capitalist Marc Andreessen recently shared his thoughts on artificial intelligence (AI) and…Ai Safety3 min readAi Safety3 min read
Max David Gupta·16 hours agoActive vs. Passive Human-AI Interaction and Effects on the Human PsycheUntil recently, human interaction with AI systems has been largely latent. We interact with AI in-built into the search engines we use every time we open Google, we scroll through movie foods on Netflix, we browse clips on Youtube. …Ai Safety2 min readAi Safety2 min read
Nick Felker·2 days agoAI is more likely to disappoint us than kill usThe opinions stated here are mine alone, not that of my employer. Last week many headlines were written and links were clicked about a topic from the Center for AI Safety. …Ai Safety8 min readAi Safety8 min read
Rafe Brena, Ph.D.inTowards Data Science·May 31Member-only4 Reasons Why I Won’t Sign the “Existential Risk” New StatementFueling fear is a dangerous game — What is the new statement? Some weeks ago, I published my pro and con arguments for signing that very well-known open letter by the Future of Life Institute — in the end, I signed it, though there were some caveats. A few radio and TV hosts interviewed me to explain what all the fuss was…Ai Safety5 min readAi Safety5 min read
Tom Sykora·May 31“AI UX Patterns”: Trustworthy AI with “fact-checking UI”One of the biggest struggles the current AI applications face is reliability —“How much can I really trust my AI’s output?” is the question all users ask nowadays. Luckily, a smartly designed interface can significantly increase your AI’s credibility in your customers’ eyes, without necessarily improving the AI quality itself. …Ai Safety3 min readAi Safety3 min read
Roman LeventovinEngineering Ideas·May 29An LLM-based “exemplary actor”Into and summary This post is the second section of “Aligning an H-JEPA agent via training on the outputs of an LLM-based “exemplary actor”, posted separately because I think it could warrant a separate discussion, largely independent of the discussion of H-JEPA agent with GFlowNet actors. …Ai Safety14 min readAi Safety14 min read
conan·May 25The Dark Side of Artificial Intelligence: What Happens When Autonomous Machines Go Rogue?Exploring the Risks and Impacts of AI Going Rogue — As we continue to push the limitations of artificial intelligence (AI), the question of what happens when autonomous machines go rogue, becomes increasingly essential. AI has brought enormous benefits to various disciplines but poses significant risks that cannot be ignored. …Ai Safety3 min readAi Safety3 min read
Richard Johnson·May 24The EU’s AI Act will change how we build softwareA Product Manager’s Perspective Do you work on software products? Do those software products include AI? If you make software, you’ve either already included AI in some format or you have it on your roadmap. AI is working its way into every product and for Product Managers, the European Union’s proposed Artificial Intelligence Act…Ai Safety10 min readAi Safety10 min read
